AVIVA RESPONSE TO LAW COMMISSION CONSULTATION ON AUTOMATED VEHICLES - Feb 2019 - AWS
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
AVIVA RESPONSE TO LAW COMMISSION CONSULTATION ON
AUTOMATED VEHICLES - Feb 2019
Aviva welcome this consultation as insurers are key stakeholders in this debate. As the UK’s largest
insurer of both individual vehicles and fleets, Aviva need to ensure our customers continue to be
protected as technological advancements progress and we are participating in some autonomous vehicle
trials. We believe this technology will bring societal benefits e.g. greater mobility giving the elderly and
disabled more independence, as well as significantly reducing road accidents caused by driver error,
leading to safer roads.
Collaboration
Aviva is working with Government directly and through the ABI and Thatcham Research, to ensure the
legislative and regulatory regimes for automated vehicles, domestically and at UNECE levels, will be fit for
purpose. Aviva contributed to the ABI response on behalf of the motor insurance industry.
Summary
• To realise the benefits of autonomous vehicles, the transition phase must be carefully managed through
appropriate legislation with focus on the safety features that underpin the technology.
• We see a clear distinction between two levels of ‘driverless cars’:
o “Assisted driving” - defining systems that provide continuous support to the driver. The driver is
required to remain engaged with the driving task and driver monitoring systems will be in place to
ensure this happens.
o “Automated driving” - systems that are capable of operating in clearly defined automated mode(s)
which can safely drive the vehicle in specified design domains without the need to be controlled or
monitored by an individual.
• We support the view that to prevent potential confusion, it is preferable to refer to “Assisted” or
“Automated” capability driving as advocated by Thatcham Research rather SAE levels.
• Defining liability following a crash will be challenging for insurers for vehicles that will be capable of
autonomy for part of a journey. These risks are inevitable, and it is imperative that vehicle
manufacturers and insurers work closely together to get the technology right to save lives.
• We support developing differing paths and requirements for Path One and Path Two Automated
vehicles/future mobility as service provisions.
• Aviva believe a clear distinction is needed between what constitutes an ’automated’ as opposed to
‘assisted driving’ vehicle (i.e. no fall back/monitoring during the automated phase by the user in
charge).
• Aviva see the need to establish, agree and embed in regulation DSSAV data minimum sets and neutral
server access to enable insurers to handle compensation claims under the Automated & Electric
Vehicle Act.
• On-going concerns for Aviva and all insurers include facing substantial difficulties in our ability to
exercise recovery rights due to limitations/shortcoming in existing Product Liability legislation and we
would favour a reverse onus of proof burden on the Vehicle Manufacturers.
• We would suggest the Law Commission consulting on the wider spectrum of vehicles that could
become autonomous which will require insurance (e.g. mining, construction and agricultural vehicles)
linking to developments in technologies as they arise to avoid unintentional risks and exposures
arising through gaps in legislation.
1
Aviva: PublicFurther contact
Aviva would be happy to discuss our response as we appreciate the pace at which ‘autonomy’ is
progressing.
Please contact: Andrew Wilkinson, Technical Claims Director, Aviva. Email:
Consultation Questions
HUMAN FACTORS
A new role in driving automation: the “user-in-charge”
Consultation Question 1):
Do you agree that:
(1) All vehicles which "drive themselves" within the meaning of the Automated and Electric Vehicles Act
2018 should have a user-in-charge in a position to operate the controls, unless the vehicle is specifically
authorised as able to function safely without one?
(2) The user-in-charge:
(a) must be qualified and fit to drive;
(b) would not be a driver for purposes of civil and criminal law while the automated driving system is
engaged; but
(c) would assume the responsibilities of a driver after confirming that they are taking over the controls,
subject to the exception in (3) below?
1. Yes, for Path One vehicles where there may be a requirement for the User in Charge within the vehicle
to re-engage at the end of a defined Automated Driving phase within a journey.
2. Path Two vehicles that are only deployed in limited local contexts (where the role of the human
occupant is that of a passenger) may still require a (potentially remote) operator to carry out
necessary surveillance operations to direct a vehicle’s actions after that vehicle has carried out a
minimal risk manoeuvre. In practice, we would envisage that these requirements may be like those
imposed on the user-in-charge.
3. We have considered if a further/alternative definition might be required, but at this stage we broadly
agree with the term user-in-charge, but it requires defined legal responsibilities. However, it is
important to emphasise that it should only apply to ‘Path 1’ vehicles that are capable of being driven
manually. For ‘Path 2’ vehicles the term ‘Operator’ is likely to be more appropriate as they will be
remote from the vehicle.
4. We do have a concern that ‘user-in-charge’ could be confusing from a consumer point of view and
that it may therefore be beneficial to explain it with a term such as ‘passive driver’ who would become
an ‘active driver’ when he/she is in control of the vehicle.
(3) If the user-in-charge takes control to mitigate a risk of accident caused by the automated driving
system, the vehicle should still be considered to be driving itself if the user-in-charge fails to prevent the
accident.
5. YES, subject to validation by timeline date-stamp in the vehicle software system to reflect when the
vehicle was in automated or manual mode. Accurate real-time data for the car is vital for this section
to operate effectively to ensure there is evidence that the driver has not interrupted a legitimate
action by the vehicle and as such caused the accident.
Consultation Question 2
We seek views on whether the label “user-in-charge” conveys its intended meaning.
6. Yes, for the reason set out above in Question 1.
Consultation Question 3
Aviva: PublicWe seek views on whether it should be a criminal offence for a user-in-charge who is subjectively aware of
a risk of serious injury to fail to take reasonable steps to avert that risk.
7. No, we disagree in making it a criminal offence for a user-in-charge to fail to take steps to avert risks
that he/she is subjectively aware of. We believe this would risk blurring the line of responsibility and
would not be conducive to road safety and public acceptance of automated driving systems.
8. The term ‘subjectively aware’ indicates that there is sufficient knowledge to form a duty of care. Not
to exercise that duty of care correctly would amount to negligence. It could be difficult to prove
awareness depending on the level of distraction/disengagement from the driving function.
9. Automated driving systems should reduce the risk of serious injury and introducing such a criminal
offence may have the unintended consequence of encouraging developers to rely on human
interventions to avoid accidents when the automated driving system should have been able to do so.
10. The situation is likely to be different if ‘conditionally automated’ vehicles were to be type approved as
these vehicles would require the driver to monitor the driving tasks and respond to any request to
take back control.
When would a user-in-charge not be necessary?
Consultation Question 4):
We seek views on how automated driving systems can operate safely and effectively in the absence of a
user-in-charge.
11. A user-in-charge may not always be required to be present in the vehicle to guarantee safety as in
likely to be the case with Path 2 vehicles. They may be operated by a licenced organisation and be
subject to requirements by local authorities to ensure that these systems are able to cope with the
scenarios that they are likely to encounter in their individual operational design domain and be able to
carry out a minimal risk manoeuvre.
12. For these systems, a remote controller would be able to substitute the user-in-charge and we would
expect the licence holder to be required to take necessary steps to ensure a safe and efficient
operation within a confined area.
Consultation Question 5):
Do you agree that powers should be made available to approve automated vehicles as able to operate
without a user-in-charge?
13. Yes. Path 2 vehicles would, at least in the foreseeable future, mainly be operated in the context of
mobility as a Service. For these vehicles, a user-in-charge who is subject to the same qualifications and
requirements as a driver would not be required.
When should secondary activities be permitted?
Consultation Question 6):
Under what circumstances should a driver be permitted to undertake secondary activities when an
automated driving system is engaged?
14. Aviva supports the insurance industry’s view that an automated driving system can safely drive the
vehicle in specified design domains without the need to be controlled or monitored by an individual.
As such, the user-in-charge should be permitted to undertake secondary activities while the
automated driving system is engaged.
15. Any vehicle that requires a human driver to act as a fall-back system is not automated and should not
fall under the definition of a vehicle that can ‘safely drive itself’ as defined in the Automated and
Electric Vehicles Act 2018. If a human driver is relied on to take back control to guarantee road safety,
the system is assisted, and the driver bears responsibility for any accidents. Drivers of these vehicles
should not be permitted to undertake any secondary activities.
16. We are concerned that a blurring of lines between a highly capable but ‘assisted driving’ vehicle and a
vehicle capable of ‘automated driving’ (i.e. not requiring the user in charge to act as a the safely
3
Aviva: Public‘backstop’) could lead to an increase in incidence where drivers might be tempted to undertake
secondary tasks. Legislation, vehicle manufacturers & insurers need to ensure consumers are clear
what can and cannot be undertaken when behind the wheel of these very different categories of
vehicles.
Consultation Question 7 (Paragraphs 3.80 - 3.96):
Conditionally automated driving systems require a human driver to act as a fall-back when the automated
driving system is engaged. If such systems are authorised at an international level:
(1) should the fall-back be permitted to undertake other activities?
(2) if so, what should those activities be?
17. No, Aviva is uncomfortable with this proposition. If the fall-back/user-in-charge is required to take
back control at short notice and has a duty of care to be aware of potential risk of serious injury, it is
difficult to see how this would be possible with any significant level of distracted behaviour being
allowed.
18. If such tasks were permitted, they must be strictly limited and controlled solely through the vehicles
own infotainment system.
CHAPTER 4: REGULATING VEHICLE STANDARDS PRE-PLACEMENT
A new safety assurance scheme
Consultation Question 8 (Paragraphs 4.102 - 4.104):
Do you agree that:
(1) a new safety assurance scheme should be established to authorise automated driving systems which
are installed: (a) as modifications to registered vehicles; or
(b) in vehicles manufactured in limited numbers (a "small series")?
(2) unauthorised automated driving systems should be prohibited?
(3) the safety assurance agency should also have powers to make special vehicle orders for highly
automated vehicles, so as to authorise design changes which would otherwise breach construction and
use regulations?
(1) regulating consumer and marketing materials?
(2) market surveillance?
(3) roadworthiness tests?
19. We understand the merits in the concept of establishing such a scheme though would question if this
could create addition burdens and duplication with existing agencies.
20. All volume produced vehicles must go through the proposed automated vehicle type approval process
framework to ensure that the whole vehicle is of an automated integrity mechanically, electrically and
electronically plus alignment to an internationally recognised standard.
21. The existing Individual Vehicle Approval (IVA) scheme should be enhanced and requirements added to
consider low volume production automated vehicles. This may require an existing authority (probably
the VCA) to be enhanced.
22. A distinction also needs to be drawn between those vehicles that operate on the public road amongst
other traffic versus those that operate on a defined and confined area.
23. We agree that unauthorised automated driving systems should be prohibited. It should be a criminal
offence to use an unauthorised automated driving system on roads or other public places.
Consultation Question 9 (Paragraphs 4.107 - 4.109):
Do you agree that every automated driving system (ADS) should be backed by an entity (ADSE) which
takes responsibility for the safety of the system?
24. Yes. In many cases, we envisage that the ADSE would be a vehicle manufacturer. Vehicle
manufacturers are already required to have a UK representative to apply to the Vehicle Certification
Aviva: PublicAgency for a certificate of conformity and to ensure that certain local requirements are met (e.g. the
direction that headlamps dip and that the speedometer displays information in miles per hour). A
similar process may be used to ensure that the automated driving system complies with local
requirements.
25. We believe that there should be a requirement for each ADSE to be locally registered and satisfy
certain capital requirements to ensure that the entity can be interacted with and pursued if it fails to
uphold necessary standards.
26. This should include enforcing minimum levels of Public Liability insurance – significant enough to
reflect the volumes of vehicles they sell and systems and process in place to maintain these, i.e. a fair
product guarantee.
Consultation Question 10 (Paragraphs 4.112 - 4.117):
We seek views on how far should a new safety assurance system be based on accrediting the developers’
own systems, and how far should it involve third party testing.
27. The vehicles must be approved in accordance either with existing international type approval
regulations, or for limited production Path 2 approved according to enhanced IVA to a national
standard. There will elements which can be self accredited in the right process and system and other
elements that requires independent assessment.
28. We believe that there should be a balance between accreditation of the developers’ own systems and
third-party testing. To prevent developers from designing to a fixed set of tests, the third-party testing
should use a randomised sample from an extensive set of test scenarios.
Consultation Question 11 (Paragraphs 4.118 - 4.122):
We seek views on how the safety assurance scheme could best work with local agencies to ensure that is
sensitive to local conditions.
29. We believe this is relevant to Path Two vehicles. A new safety assurance authority would need to work
with the relevant local agencies responsible for the roads on which these vehicles are to operate. it
would be preferable to have a single authority who provides both safety assurance and licence to
operate all Path two vehicles as far as is reasonable possible.
30. However, regardless of whether they are Path 1 or Path 2 vehicles, they should conform with a
minimum set passive safety features to keep the driver and passengers protected within the vehicle
from various crash forces. The precise requirements are likely to vary depending on the automated
driving system’s operational design domain and local traffic conditions.
CHAPTER 5: REGULATING SAFETY ON THE ROADS
A new organisational structure?
Consultation Question 12 (Paragraphs 5.30 - 5.32):
If there is to be a new safety assurance scheme to authorise automated driving systems before they are
allowed onto the roads, should the agency also have responsibilities for safety of these systems following
deployment?
If so, should the organisation have responsibilities for:
(1) Regulating consumer and marketing materials?
(2) Market surveillance?
(3) Roadworthiness tests?
We seek views on whether the agency’s responsibilities in these three areas should extend to advanced
driver assistance systems.
31. We believe that automated driving systems should undergo rigorous testing to ensure that they are as
safe as possible before they are authorised to be used on roads or other public places. An approval
regime based on both self-certification and third-party testing should evolve to ensure the safety of
new automated driving.
5
Aviva: Public32. The existing MOT structure needs to be maintained and refined to cater for the increasing technology
and electronics to include a check of such systems. The organisation tasked with safety-related
aspects of automated driving systems needs to set clear standards that need to be met by the vehicle
(passive safety features) and the automated driving system (hardware and software).
33. Providing consumers with marketing materials, education and training to develop and maintain
awareness of their responsibilities when using an automated driving system will be a significant factor
to ensure road safety.
Driver training
Consultation Question 13 (Paragraphs 5.54 - 5.55):
Is there a need to provide drivers with additional training on advanced driver assistance systems?
If so, can this be met on a voluntary basis, through incentives offered by insurers?
34. Yes, we agree that such training is required but do not necessarily agree that insurers are the right
organisations to incentivise this in isolation.
35. Insurance premiums will reflect the risk and it may be that insurers will consider a driver who has
undertaken such training to be a better risk and therefore offer incentives in the way of premium
discounts. However, this would require the full co-operation of motor manufacturers to provide data
to help us analyse how risk is affected by such training. It is also possible that such training would
need to be repeated/updated as the reduction in risk through such training may be effective for a
limited time.
36. We would suggest there may also be other areas where this could be implemented, i.e. as part of the
driving test for drivers intending to use vehicles with an element of autonomy.
Accident investigation
Consultation Question 14 (Paragraphs 5.58 - 5.71):
We seek views on how accidents involving driving automation should be investigated.
We seek views on whether an Accident Investigation Branch should investigate high profile accidents
involving automated vehicles? Alternatively, should specialist expertise be provided to police forces.
37. Initially there will be a limited number of automated vehicles on the road and it is important for
consumer confidence to track crashes. In depth analysis of specific crashes will depend on their nature
and the potential consequences.
38. Police Forces already have considerable expertise, resources and capability to undertake vehicle
accident investigations. While a dedicated Instigation branch may have attractions, we would not
wish this to dilute the existing Police resource.
39. We believe the existing authorities are well placed to investigate causes of accidents, including
obtaining data from autonomous vehicle manufacturers and through appropriate links to VCA, DVSA.
The VM’s can feedback to avoid the same issue being the cause of further accidents.
40. However, the police have a duty to gather evidence to prosecute on behalf of victims. If an
autonomous vehicle is at fault, there may not be a driver to prosecute, but depending on whether any
criminal liability attaches to the owner or manufacturer, there may still be a need for police
involvement.
Setting and monitoring a safety standard
Consultation Question 15 (Paragraphs 5.78 - 5.85):
The technical challenges of monitoring accident rates
(1) Do you agree that the new safety agency should monitor the accident rate of highly automated
vehicles which drive themselves, compared with human drivers?
41. Yes. It is essential that the causes and consequences of different vehicles types, driving scenarios and
conditions is recorded, measured and analysed. This will identify the risk factors which need to be
managed and the effectiveness of existing risk management measures including regulation, legislation,
Aviva: Publiceffectiveness of enforcement etc. This learning will be important for vehicle manufacturers to
continue to develop the technology as well as for insurers to improve our understanding of risk and be
able to price accordingly.
42. The proposed safety agency would be well placed to carry out this. Monitoring the number and
severity of accidents involving autonomous vehicles could provide valuable information about the
driving system as well as the interaction with other road users.
43. It would seem sensible for Dept for Transport to collect this data as part of their statistics on reported
accidents – Stats 19 to facilitate a comparison.
44. The key requirement will be for automated vehicles to report any accident in which they are involved
and to send a ‘crash alert’ message to the insurer concerned. The message will need to contain a
relatively small set of data (DSSAV), which is necessary for insurers to establish liability.
45. We therefore believe that there is merit in exploring a statutory requirement to collect, store and
transfer DSSAV to a neutral server. While we would hope that such a requirement is introduced via
the type approval process, the Law Commission should explore the option of amending the
Automated and Electric Vehicles Act 2018 to make this mandatory.
(2) We seek views on whether there is also a need to monitor the accident rates of advanced driver
assistance systems.
46. Insurers will need to monitor, interpret and reflect on these accident rates to be able to price risk
accordingly. It is important that any vehicle with autonomous capabilities is identified to the insurer at
the outset. The sharing of data held by the ADSE or the vehicle manufacturer will be critical to
insurers.
Consultation Question 16:
(1) What are the challenges of comparing the accident rates of automated driving systems with that
of human drivers?
47. There challenges in comparing automated driving & human driver accidents including, a current
dependency on specialists or VMs to access vehicle data, no enforced mandatory reporting of non-
injury accidents.
48. There will not be sufficient volumes of data on autonomous vehicles involved in accidents for some
considerable period to make meaningful data comparisons. Where accidents involve more than one
vehicle and only one of those vehicles was driving in autonomous mode, it will need to be decided
how this should be reflected in such statistics.
(2) Are existing sources of data sufficient to allow meaningful comparisons? Alternatively, are new
obligations to report accidents needed?
49. The insurance industry has long dealt with data to analyse frequency and risk so will be well placed to
interpret and understand the relative accident rates and the cost of accidents of vehicles fitted with
advanced driver assistance systems. However, it is currently difficult to establish on which vehicles they
are fitted, particularly where the customer has specified them as an individual add-on. This is an area
where the sharing of data held by the ADSE or vehicle manufacturer is critical, and we would welcome
the Law Commission exploring making access to in-vehicle data more generally as part of its ongoing
review.
CHAPTER 6: CIVIL LIABILITY
Is there a need for further review?
7
Aviva: PublicConsultation Question 17 (Paragraphs 6.13 - 6.59):
We seek views on whether there is a need for further guidance or clarification on Part 1 of Automated and
Electric Vehicles Act 2018 in the following areas:
(1) Are sections 3(1) and 6(3) on contributory negligence sufficiently clear?
50. We support the aims of the A&EV Act and share the Government’s objective of making it as easy as
possible for consumers to understand what insurance cover they need; and ensuring that any injured
parties (including users of these vehicles) have their claims settled quickly. As motor insurers we are
committed to providing the cover that protects those who take up this technology and ensure that other
road users are protected.
51. We would welcome further clarity on section 3(1) of the Act as the explanation is difficult to follow in
terms of comparing the claim to a claim ‘brought by the injured party against a person other than the
insurer or vehicle owner’. It is difficult to know exactly what scenarios are envisaged here. We believe
this could be explained much more simply, as has been done in S.6.3 of the Law Commission Summary
of the Preliminary Consultation Paper.
(2) Do you agree that the issue of causation can be left to the courts, or is there a need for guidance
on the meaning of causation in section 2?
52. It would be helpful to have further guidance on when an accident is deemed to be ‘caused’ by the
autonomous vehicle, especially in respect of the situation set out in Q.1(3) where there may be
ambiguity over whether the vehicle or user-in-charge has caused the accident. This could be left to
the courts but evidentially it could be difficult to ascertain and as such further guidance around this
would, in our view, be helpful.
(3) Do any potential problems arise from the need to retain data to deal with insurance claims? If so: (a)
to make a claim against an automated vehicle’s insurer, should the injured person be required to
notify the police or the insurer about the alleged incident within a set period, so that data can be
preserved?
53. Insurers will require sufficient data to allow them to comply with the AEV Act, i.e. to ascertain for
certain which mode the vehicle was in at the time of the accident. This needs to be available to
insurers and a clear mechanism agreed as to how it will be provided.
(b) how long should that period be?
54. S.170 of the Road Traffic Act covers the current obligation to stop after an accident and provide details
and where relevant to notify the police within a reasonable time or in any case within 24 hours. The
section does refer to ‘drivers’ and ‘mechanically propelled vehicles’, both of which may need
amendment to cover future automated and electric vehicles. The period for retention of data would
need to allow sufficient time for requests to be made by police/insurers after notification.
Civil liability of manufacturers and retailers: Implications
Consultation Question 18 (Paragraphs 6.61 - 6.116):
Is there a need to review the way in which product liability under the Consumer Protection Act 1987
applies to defective software installed into automated vehicles?
55. We believe that there is a need to reform the UK’s product liability regime. The Consumer Protection
Act (CPA) currently only covers embedded software as it is contained in the manufactured goods.
There is currently no direct mention of software or digital content.
56. The current Act states it is necessary to prove that a product is defective. As automated driving
systems are complex and accessing the data from the VM to prove there was a defect may restrict the
insurer being able to claim against the VM or software provider.
Aviva: Public57. The reversal of the burden of proof where a vehicle is in autonomous mode would encourage
manufacturers to share the relevant vehicle data. This would act as a deterrent to bringing a system
on to the road where safety concerns have been identified during the development stage.
58. The vehicle manufacturer and software provider may not be the same organisation and software will
need to be regularly updated/monitored and as such the definition of ‘producer’ and ‘product’ may
need to be reviewed in the CPA.
Consultation Question 19 (Paragraphs 6.61 - 6.116):
Do any other issues concerned with the law of product or retailer liability need to be addressed to ensure
the safe deployment of driving automation?
59. Yes. The UK Road Traffic Act requires insurers to provide unlimited cover for death and personal injury
whilst limits are placed on liability for other areas of loss such as liability to property or goods. This
can result in a recovery short fall or gap where an autonomous vehicle is ultimately found liable.
Motor premiums would need to reflect this.
60. We would highlight that insurers would need to be able to recover other current statutory recovery
costs from automated vehicle manufacturers or suppliers e.g. NHS costs, recovery and storage fees.
CHAPTER 7: CRIMINAL LIABILITY
Offences incompatible with automated driving
Consultation Question 20 (Paragraphs 7.5 - 7.11):
We seek views on whether regulation 107 of the Road Vehicles (Construction and Use) Regulations 1986
should be amended, to exempt vehicles which are controlled by an authorised automated driving system.
61. Regulation 107 will require amendments to allow for fully automated vehicles operating on public roads
without a driver and without a centralised remote-control capability. We would anticipate most
autonomous mobility on demand services to operate in this way (Path Two Vehicles).
Consultation Question 21 (Paragraphs 7.5 - 7.11):
Do other offences need amendment because they are incompatible with automated driving?
62. There are various other provisions under the Road Vehicles (Construction and Use) Regulations 1986
which are incompatible with autonomous vehicles. For example:
63. Part 1 S.3 Provides definitions of ‘bus’, ‘mini-bus’ and ‘pedestrian controlled vehicle’ all of which
reference a ‘driver’ which may not be relevant for autonomous vehicles.
64. Under Part 2 there are various provisions regarding such things as vision, speedometers, audible
warning instruments, seatbelt anchorage points (containing reference to ‘driver’), parking in darkness,
drivers control, reversing and television sets, all of which may be irrelevant or need amendment in the
case of full autonomy.
65. Regulation 104 may also require amendment but probably only to clarify what is meant by ‘road and
traffic ahead’. If the automated vehicle has turned into a blind alley and its only way out is to reverse
back onto the main road and it needs the intervention of a remote operator to do so; does the operator
require a clear view of the road and traffic ahead or, in this case, is it the road and traffic behind?
Offences relating to the way a vehicle is driven
Consultation Question 22 (Paragraphs 7.14 - 7.19):
Do you agree that where a vehicle is:
(1) listed as capable of driving itself under section 1 of the Automated and Electric Vehicles Act 2018; and
(2) has its automated driving system correctly engaged;
the law should provide that the human user is not a driver for the purposes of criminal offences arising
from the dynamic driving task?
66. Agree. Careful consideration needs to be given to the duty of care placed on a user-in-charge to avoid
potential serious injury and how far, if at all, this extends to avoiding criminal offences (such as
speeding, illegal parking etc). The answer to this question may be different for vehicles where there is
a user-in-charge (Path One) in the vehicle and for those fully autonomous vehicles that do not require
a user-in-charge in the vehicle (Path Two) as was referred in the response to Q1.
9
Aviva: PublicConsultation Question 23 (Paragraph 7.21):
Do you agree that, rather than being considered to be a driver, a user-in-charge should be subject to
specific criminal offences? (These offences might include, for example, the requirement to take
reasonable steps to avoid an accident, where the user-in-charge is subjectively aware of the risk of serious
injury (as discussed in paragraphs 3.47 to 3.57)).
67. Yes. This question also highlights the need for very clear rules for the capability of Automated vehicles.
If the Automated vehicle is not capable of doing a better job than a human driver in terms of speed
and appropriateness of response, the vehicle should not be classed as automated as was referenced in
our response to question 3.
68. We believe this is a similar situation to our current common law duty of care. We suggest that
different people deal differently based on their own personal levels of competency, experiences and
training and no two situations will be the same.
Consultation Question 24 (Paragraphs 7.23 - 7.35):
Do you agree that:
(1) A registered keeper who receives a notice of intended prosecution should be required to state if
the vehicle was driving itself at the time and (if so) to authorise data to be provided to the police?
69. Yes. We believe the registered keeper should be entitled to free access to the vehicle data from the
Vehicle manufacturer to prove this.
Where the problem appears to lie with the automated driving system (ADS) the police should refer the
matter to the regulatory authority for investigation?
70. Yes. Mandatory reporting should be required by the VM to remain on the Automated list and free
from sanctions. The Police would need the resources to action this and consideration should also be given
to the process for appeals etc to demonstrate a robust system is in place.
(2) Where the ADS has acted in a way which would be a criminal offence if done by a human driver,
the regulatory authority should be able to apply a range of regulatory sanctions to the entity behind the
ADS?
71. Yes, we agree.
(3) The regulatory sanctions should include improvement notices, fines and suspension or withdrawal
of ADS approval?
72. Yes, agreed but more details on these proposals would be required including any impact on the safety
of the current owner / user.
Consultation Question 25 (Paragraphs 7.37 - 7.45):
Do you agree that where a vehicle is listed as only safe to drive itself with a user-in-charge, it should be a
criminal offence for the person able to operate the controls (“the user-in-charge”):
(1) Not to hold a driving licence for the vehicle.
(2) To be disqualified from driving.
(3) To have eyesight which fails to comply with the prescribed requirements for driving;
(4) To hold a licence where the application included a declaration regarding a disability which the user
knew to be false?
(5) To be unfit to drive through drink or drugs: or
(6) To have alcohol levels over the prescribed limits.
73. Yes, we agree with all these proposals. Where a ‘user in charge’ is required they must be fit and
capable to meet the minimum UK road legislation requirements – when needed to assume their driving
responsibilities.
Aviva: PublicConsultation Question 26 (Paragraphs 7.37 - 7.45):
Where a vehicle is listed as only safe to drive itself with a user-in-charge, should it be a criminal offence to
be carried in the vehicle if there is no person able to operate the controls.
74. Yes.
Responsibilities for other offences
Consultation Question 27 (Paragraphs 7.48 - 7.65):
Do you agree that legislation should be amended to clarify that users-in-charge:
(1) Are ‘users’ for the purposes of insurance and roadworthiness offences; and
75. Yes
(2) Are responsible for removing vehicles that are stopped in prohibited places, and would commit a
criminal offence if they fail to do so?
76. We agree with this proposal. For vehicles that are authorised to operate without a user-in-charge,
these requirements should fall onto the ADSE or operator.
Consultation Question 28 (Paragraphs 7.59 - 7.61):
We seek views on whether the offences of driving in a prohibited place should be extended to those who
set the controls and thus require an automated vehicle to undertake the route.
77. We would expect automated driving systems to be programmed in a way that does not allow them to
be used in prohibited places.
78. We are, however, aware of some vehicle manufacturers’ intention to develop autonomous off-road
capability and this is likely to make geo-fencing more difficult. For these types of vehicles, the user-in-
charge may ultimately have to bear responsibility for any associated offences.
Obligations that pose challenges for automated driving systems
Consultation Question 29 (Paragraphs 7.71 - 7.88):
Do you agree that legislation should be amended to state that the user-in-charge is responsible for: (1)
duties following an accident;
(2) complying with the directions of a police or traffic officer; and
(3) ensuring that children wear appropriate restraints?
79. Yes. We agree with this proposal. For vehicles that are authorised to operate without a user-in-charge,
these requirements should fall onto the ADSE or operator.
Consultation Question 30 (Paragraphs 7.71 - 7.88):
In the absence of a user-in-charge, we welcome views on how the following duties might be complied with:
(1) duties following an accident;
(2) complying with the directions of a police or traffic officer; and
(3) ensuring that children wear appropriate restraints?
80. We believe the remote operator, who takes the place of a user-in-charge, would need to comply with
(1). The remote operator would also need to be able to recognise directions of a police or traffic officer if
fully autonomous. As for the wearing of seatbelts of passengers in fully autonomous vehicles, we would
not expect such a vehicle to be able to commence its journey until the passengers had complied with any
restraint protocols.
Consultation Question 31 (Paragraphs 7.71 - 7.88):
We seek views on whether there is a need to reform the law in these areas as part of this review.
81. Yes, reforms will need to be made to reflect any new requirements in these areas as outlined above.
Aggravated offences Consultation Question 32 (Paragraphs 7.92 - 7.123):
We seek views on whether there should be a new offence of causing death or serious injury by wrongful
interference with vehicles, roads or traffic equipment, contrary to section 22A of the Road Traffic Act
1988, where the chain of causation involves an automated vehicle.
11
Aviva: Public82. We agree that a new offence is needed. It seems sensible to specifically explore this issue in relation to automated driving systems and clarify the legal position. Consultation Question 33 (Paragraphs 7.113 - 7.123): We seek views on whether the Law Commissions should review the possibility of one or more new corporate offences, where wrongs by a developer of automated driving systems result in death or serious injury. 38 83. Yes. New corporate offences that hold individual managers or directors, as well as the organisation, to account should be considered. Such offences would be particularly appropriate where it is evident that an organisation’s culture has led to the systematic neglect of safety standards and resulted in death or serious injury. However, there would need to be consideration as to how state of the art defence may be used, or this could have limited value. CHAPTER 8: INTERFERING WITH AUTOMATED VEHICLES Consultation Question 34 (Paragraphs 8.1 - 8.58): We seek views on whether the criminal law is adequate to deter interference with automated vehicles. In particular: (1) Are any new criminal offences required to cover interference with automated vehicles? (2) Even if behaviours are already criminal, are there any advantages to re-enacting the law, so as to clearly label offences of interfering with automated vehicles? 84. While the described behaviours are already criminal offences, we believe that it would make sense to create specific offences to act as a deterrent. These could be framed as ‘(causing death by) interference with the lawful operation of automated driving systems’. It would seem sensible to clarify the law to ensure that all unlawful captured by existing or new offences. 85. With functionality increasingly being driven through software, there should be a new offence to address the bypassing and ‘cracking/hacking’ of software by users. Tampering with vehicles Consultation Question 35 (Paragraphs 8.28 - 8.31): Under section 25 of the Road Traffic Act 1988, it is an offence to tamper with a vehicle’s brakes “or other mechanism” without lawful authority or reasonable cause. Is it necessary to clarify that “other mechanism” includes sensors? 86. Yes, and suggest this should include wider technical aspects e.g. software, programming, algorithms and communications equipment. Unauthorised vehicle taking Consultation Question 36 (Paragraphs 8.32 - 8.39): In England and Wales, section 12 of the Theft Act 1968 covers “joyriding” or taking a conveyance without authority, but does not apply to vehicles which cannot carry a person. This contrasts with the law in Scotland, where the offence of taking and driving away without consent applies to any motor vehicle. Should section 12 of the Theft Act 1968 be extended to any motor vehicle, even those without driving seats? 87. Yes. We believe a consistent approach across the whole of the UK would be safest. Causing danger to road users Consultation Question 37 (Paragraphs 8.6 - 8.12): In England and Wales, section 22A (1) of the Road Traffic Act 1988 covers a broad range of interference with vehicles or traffic signs in a way which is obviously dangerous. In Scotland, section 100 of the Roads (Scotland) Act 1984 covers depositing anything a road, or inscribing or affixing something on a traffic sign. However, it does not cover interfering with other vehicles or moving traffic signs, even if this would raise safety concerns. Should section 22A of the Road Traffic Act 1988 be extended to Scotland? 88. Yes, it should apply to Scotland. Aviva: Public
CHAPTER 9: “MACHINE FACTORS” – ADAPTING ROAD RULES FOR ARTIFICIAL INTELLIGENCE DECISION-
MAKING
Rules and standards
Consultation Question 38 (Paragraphs 9.6 - 9.27):
We seek views on how regulators can best collaborate with developers to create road rules which are
sufficiently determinate to be formulated in digital code.
89. The priority for insurers is that road rules can be applied equally to human drivers and automated
driving systems. There should be no separate Highway Code or obligations depending on the driving mode
as this could lead to confusion among road users and be detrimental to road safety.
90. The investment in road furniture, markings and surface maintenance all need consideration when
addressing the need to accurately and safely formulate a digital view of the environment. Initial
automated vehicles will require “roads cars can read” and therefore an investment in infrastructure (white
lines) etc will be a prerequisite to enable the use and safe function of automated systems.
Consultation Question 39
We seek views on whether a highly automated vehicle should be programmed so as to allow it to mount
the pavement if necessary:
(1) to avoid collisions;
(2) to allow emergency vehicles to pass;
(3) to enable traffic flow;
(4) in any other circumstances?
91. The same rules should apply to human drivers and automated driving systems. Our preferred approach
would be to clarify under which circumstances it would be not be lawful to mount the pavement.
At least initially, autonomous mode should only be used in specified areas, so we would envisage the user
in charge making an informed judgement.
Consultation Question 40 (Paragraphs 9.6 - 9.37):
We seek views on whether it would be acceptable for a highly automated vehicle to be programmed never
to mount the pavement.
92. We refer to our response to Question 39.
Should highly automated vehicles ever exceed speed limits?
Consultation Question 41 (Paragraphs 9.40 - 9.47):
We seek views on whether there are any circumstances in which an automated driving system should be
permitted to exceed the speed limit within current accepted tolerances.
93. No. An automated vehicle should only operate within the law. In time technology may allow for
variable speed limits dependent on traffic density on connected highways to be fed in real time to such
vehicles.
Edging through pedestrians
Consultation Question 42 (Paragraphs 9.49 - 9.55):
We seek views on whether it would ever be acceptable for a highly automated vehicle to be programmed
to “edge through” pedestrians, so that a pedestrian who does not move faces some chance of being
injured. If so, what could be done to ensure that this is done only in appropriate circumstances?
94. Yes. Human drivers should only ever edge forward into safe, i.e. unoccupied, space in front of the
vehicle. This will sometimes mean only moving an inch or two at a time to avoid touching a pedestrian.
95. An automated vehicle could be programmed to do the same, however an automated vehicle
aggressively nudging its way through pedestrians would be unacceptable.
Avoiding bias in the behaviour of automated driving systems
Consultation Question 43 (Paragraphs 9.68 - 9.74):
To reduce the risk of bias in the behaviours of automated driving systems, should there be audits of
datasets used to train automated driving systems?
13
Aviva: PublicN/A
Transparency
Consultation Question 44 (Paragraphs 9.76 - 9.88):
We seek views on whether there should be a requirement for developers to publish their ethics policies
(including any value allocated to human lives)?
96. In the absence of a global agreed standard, yes.
97. We should not expect automated driving systems to make ethical judgments. These systems should be
safe enough to avoid accidents in most circumstances. If they are faced with a situation where an accident
is unavoidable, we would not expect the system to be capable of making such decisions.
Consultation Question 45 (Paragraphs 9.76 - 9.88):
What other information should be made available?
N/A
Future work and next steps
Consultation Question 46 (Paragraphs 9.91 - 9.93):
Is there any other issue within our terms of reference which we should be considering in the course of this
review?
98. Access to vehicle data:
• A critical area for insurers is availability of in-vehicle data immediately following a crash. We believe
that a statutory requirement to collect, store and transfer this data should be explored.
• Statistics relating to the number of accidents caused or contributed to by the ADS;
• Full and detailed disclosure of the capabilities of an ADS which should include the SAE level of that
vehicle; and
• An extensive description of any update to an ADS including the improvements brought and whether
the update is safety critical or not.
99. Vehicle list and standards:
The Automated & Electric Vehicles Act is underpinned by the creation of a list by the Secretary of State for
vehicles capable of safe and lawful driving. For the protection of consumers of automated vehicles, other
road users, the public and insurers, it is imperative that technical standards clearly defining the minimum
requirements are developed and form the basis of Type Approvals. These standards are likely to
determine if a specific vehicle type is placed on the list in the UK.
Work towards this goal have been progressing at international level (GRVA). Such work should not be
allowed to stall or be diluted. We call upon the Law Commission to work with Government and its
agencies to continue development of such international standards on the current ‘ground up approach’
and resist efforts to adopt a ‘top down approach’ leaving it to Vehicles Manufacturers to create their own
safety standards and self accreditation as they develop and refine automated driving products.
In the absence of international safely standards being applied consistently by all VMs, there will be
increased pressure and propensity to include vehicles on the Secretary of States list whose capabilities fall
short of what motor insurers would consider a safe automated driving system.
100. Components/equipment
• Identification of component manufacturer for AV’s will be an issue.
• For small component manufacturers – e.g. capacitors, resistors, CPU’s, this may be difficult to identify
and original VM’s may need to take collective responsibility for any AV system failure.
• VM’s may be obstructive when accident investigations require diagnostic evidence, unless such
evidence, triggered by ‘crash alert accelerometers’ telematic parameters are automatically sent to the
Cloud.
Aviva: Public• Likelihood of subrogation against VM or component manufacturer. More likely where diagnostic
equipment is robust, and data retained in the event of an accident and more importantly,
independent analysis is carried out to identify the nature of the failure. This assumes standardised
systems development and diagnostic abilities.
101. Insurance limits
• Issue over levels of insurance cover as between motor unlimited and PL/Products manufacturers in
the event of a case where the VM subrogates against the VM or component/software manufacturer.
• Claimants unlikely to be fully compensated if subrogation successful but products liability insurance
limits are inadequate. Additional burden on VM’s and component manufacturers to buy unlimited or
significantly higher limits.
• Pricing for higher products liability limits may be more difficult in the absence of claims data.
• If higher limits are stipulated, does this just apply to VM’s or to any component manufacturers which
have products knowingly incorporated into automated vehicle systems?
102. Quality Assurance
• Component manufacturers should be required to meet safety assurance systems and accreditation
where incorporated into original ADS systems as well as aftermarket repairs/replacement.
• Type approval should include consideration of whole vehicle integrity to incorporate AV systems.
• Tamper proof systems to prevent non-accredited aftermarket repairs.
• How is software updated and by who? It is important that the integrity of the software/hardware is
ensured to a standard that protects. It not only from internal tampering within say motor trade
environment for altering vehicle speeds allowed, but also from malicious external attacks such as
hackers etc. This will be exacerbated by the plugins /apps which can now link to a vehicle.
• In vehicle data should be made available to ensure users receive redress in the event of an accident –
statutory requirement to supply to the relevant parties and within a prescribed time? Tesla, in
particular, is protective of their crash data to protect brand reputations and ability to investigate.
• The use of level 3 systems poses the greatest threat and there should be additional safeguards in place
to ensure that vehicles are always adequately controlled.
• Training provided by VM to original purchaser – subsequent purchasers? When does the
responsibility of the VM end for both training and hardware/software updates?
• Is software considered a product even if intangible? If defective, what are the rights of redress under
current legislation?
• Opportunity for hacking or aftermarket updates by specialists to override parameters of VM?
103. Parallels with Aviation and Maritime industries
Aviva welcomes the parallels drawn with aviation and maritime industries with similar challenges in the
transition from assisted driving to full autonomy where automated navigation is already common. We
suggest this could be explored in a further consultation particularly on:
• Researching human behaviours e.g. over reliance on the automated systems to assist decision making,
complacency of systems, pilots attentiveness (or lack of), and relative experience of pilots.
• Use of black box or flight recorders.
• Foreign jurisdictions and learnings from authorities in other countries that provide a single
body/approach over road, rail, marine and aviation accidents.
• Learnings from aviation on pre-determined routes for fully autonomous vehicles that lack any driver
controls and communicate with one another. For example, continual assessment of risks as AV’s
develop, learning from mistakes plus further testing as it will not be possible to identify all the issues
initially.
End
15
Aviva: PublicYou can also read
